1
1 Comment

How to monetize our open source tools?

smartcat-labs/ranger

We at smartcat.io have created and open-sourced data and load generator tools:
Ranger contextual data generator
Berserker flexible load generator
(Please give the repositories a GitHub star if you feel they deserve it).

They allow us to test the data pipelines and help us in our day to day consulting work. We've got lots of positive feedback on the tools, especially for the Ranger, data generator. To give you a context, Ranger allows us to declaratively, in a YAML config file, specify how our data looks to be as production-like as possible. For example, it lets us declare following:
Create 1 000 000 users, whose first names are from this set {Alice, Bob, Charlie, ...} , surnames from this set { Alison, Blair, Carlson, ... }, born between 1970 and 1999, and so on. This declaration creates 1M users (or any other entity you specify) randomly based on the values. However, it also allows for following: "Out of these 1M users, let exactly 100 be born in 1986". The latter feature allows us to test complex queries and data transformations.

So, it is quite flexible and useful as it is now, free and open source, for us and other engineers/companies which use it.

However, I was wondering if it is possible to monetize this somehow. I could not think of a viable business model for these tools. Monetization would allow us to develop and maintain the tools actively, instead of weekends and in between our consulting gigs.

Do you have any ideas how to monetize these tools?

EDIT:
Some thoughts/bullets on the tools:

  • Used for data pipelines / system performance testing
  • Used (mostly) by data engineers
  • Maybe it could be used for training ML algorithms (generating test data, or at least some noise data)
  • Probably could be used in large companies with large data pipelines
  • Ranger could be used as Java library for unit/integration testing
  1. 1

    <joking>Wrap it up behind a website and use it to generate fake identity data?</joking>

    The filtering + data transformation verification part sounds interesting. Maybe whip that up under a QuickCheck-like random property verification ("for all random sets where exactly 100 users are born in 1986, there are still 100 users born in 1986 after running the transformation"), and see if it actually catches bug in peoples pipelines.

    Could let people donate if it found bugs.